(Disclaimer: I imagine that Hubitat surely doesn't recommend this, but that doesn't mean people don't do it)
Do any of you use Hubitat in high consequence applications?
For example, in a scenario where a frozen system or incorrect automation would directly or indirectly result in costly damage to something?
Would you be willing to share more about your application and experience?
I'm trying to determine how much to trust Hubitat and local Zigbee / Z-Wave based home automation in general. Would love to learn more anecdotally through you all : )
I use mine for flood prevention & home security, although Hubitat has & does NOT reccomend it as a security or other life critical device, well because their lawyers told them so. In over 6 years I've had a few hiccups, unable to turn off sirens remotely, cops called my daughter(long story). Or I ( that's on ME) let the battery die in a flood sensor or two, although no resultant damage. Overall I'd say my system is on par with my sister's ADT security system, which has also brought the police unneccesarily. I trust my system enough that I have battery backup on both hubs( 1 hub for flood, 1 for security) and feel confident enough to not need a professional system. I would say if you use the hub for something important, that's your decision, but I'd advise minimizing the load on the hub running critical apps.
Majority of my security sirens are battery or have battery backup, so even in a power outage my security still works, for about 48 hours(PIPower Unit dual 21700 LiIon Cells~~9000mAh)
Personally, I consider "security or other life critical device" as VERY different use cases. I don't have a security system. I do have HE doing leak detection and turning off a master water valve. Most people in the world, especially outside of home automation have neither of those. So while I don't want it to fail, and I have UPSes, alerting, etc. - I could live if my HE failed in a security or water leak situation.. - Not a preferred scenario, but life will go on.. (despite costly damage potential - but I do have insurance). I still change internal sink and washer hoses on a 10 year cycle (and prefer copper/hard pipe fittings).
That all said, "life critical" is an entirely different matter, IMHO - I consider that like a T-Stat to supply heat, or maintaining O2 levels in an enclosed space. - I would NEVER trust HE (nor HA, ST, Homey, etc.) for such a task. If people can die, then a dedicated purpose built device is needed - I personally, think even gas fireplace automation is on the cusp of the "safe/not safe" edge.
You COULD built a T-Stat with sensors, and discrete relays to enable a gas valve in a furnace, and monitor for over-temperatures, combustion air positive pressure, etc. - I would never do that, that is why people build and test Thermostats. - I'm fine if those have interfaces for HE to monitor and perhaps do supervisory control, but "mission critical" safety systems (that we use for industrial applications at my work, that manage toxic chemicals and high energy/thermal processes) are built with redundancy and fail-safe modes, and very little of that is wireless and most of it has redundant wiring running over different routes
So monitoring, and alerting for "security, safety & important events" (more that turning on lights), that all seems OK by me. "Human Safety" and direct control of critical systems with HE, that seems like a use-case to far, by my instincts.
I have 4 production Hubitat hubs that I use to automate absolutely everything that can be automated in my home, and it has been very reliable. That said, none of those systems is life critical, and they do on occasion fail. That isn’t an issue since their automation is for convenience and prevention and I have things setup in such a way that everything can also be done manually.
I wouldn’t recommend using this system for a high consequence application. It is not what it is built (and costed) to do. For those systems, you need something that is 100% reliable and as such includes redundancy/failover options.
I feel like maybe this is mining for data VS an actual question.
A person can use zwave devices on a certified life safety alarm system like ones from Qolsys if they chose. Those are monitored and certified for this so insurance discounts etc are possible.
I've mentioned this in quite a few threads now, since it seems to touch so many topics, but I have been running my furnace with Hubitat acting as the thermostats for my third winter now.
The high consequence is that if Hubitat fails in sub zero weather, my copper heating pipes will freeze.
However, I have actually had zero issues with the system. I use virtual thermostats and Zigbee temp sensors, which feed a controller app. The controller app feeds temps to the virtual thermostats, and it reacts to the operating state of the thermostats to turn on a relay on a 4 channel relay board. That relay sends 24v AC to the zone valve for that thermostat, until the virtual thermostat goes idle, then the controller will turn off the relay and cut power to the zone valve.
I built in safety features.
I still have physical Honeywell T6 Zwave thermostats for each zone. They still have power, and they connect to a 4-channel non-zigbee 24V relay board. The non-zigbee board switches if the furnace is connected to the physical thermostats, or to the Zigbee board using the virtual thermostats. I control three zones, so the extra relay on the zigbee board is used to send power to the control signal inputs of the non-zigbee board, to flip them all to use the Zigbee board. That way the powered down state is to just use the thermostats. If I can't press the zigbee disengage relay in Hubitat, there is a button on the Zigbee board to manually return everything to thermostats. I've never had to push it. Knock on wood.
I have methods that check periodically if the state of the zone valves does not match what it should be for the operating state of the virtual thermostats. I've also turned on command retry now that that is available, in case a command has missed. Still hasn't happened...
I monitor the Zigbee temp sensors batteries, by reported value and if they have reported in the last six hours. If not I get alerts. Preemptively changing the batteries helps too, to never have one die.
I also have my controller sync the virtual thermostats to the physical T6s, so the T6s basically become input and display terminals when Hubitat is in control, so to anyone who doesn't know, it still seems like the T6s are in control, but it is all Hubitat and Groovy under the hood.
Edit: Oh, I do all this because all smart thermostats that I've owned are horrible at controlling actual temperature, so I wrote myself a virtual cycling anticipating thermostat driver to get temp swings under control.
you got me! i am data mining!... to learn from others and to determine how much to trust my Hubitat and associated devices for some important things around my house.
i know that i wouldn't trust my Sonos for even the least critical of applications, for example. my own anecdotal evidence there is ample. but i'm new to the Hubitat and Zigbee/Z-Wave platform so i don't have my own mental model built up yet about when/how/why it might fail. thus, my crowdsourcing from this (amazingly helpful!) forum
We all are hardly a randomized sample, so just noting you're going to get data biased heaviiy towards power users w/maybe some moderate level/newbs sprinkled in who happened to visit near when your topic was posted.
This issue has been discussed ad nauseum (and particularly recently, IIRC), a little searching on your part will provide a trove of answers from the community in that topic - that discussion was somewhat more focused on alarm system use so the general terms like security system, alarm system, etc., could be a good place to start w/your keyword searches. NVM, only took a minute to find it:
As noted it was alarm system focused, but very relevant to this topic.
1 - If it CAN fail, it WILL
2 - Nothing cannot fail, reread #1
3 - Anything used for high consequence has redundancies, some things double redundancy.... even that fails occaisionally , reread #1 and #2
Plan for 1 through 3. If you are using a system designed for convenience in a high consequence scenario, make a redundant failsafe (or two). If you have a redundant failsafe, have something external to both systems to alert you that one or both systems failed and a plan for what to do if that happens. Reread #3.
One minor point, around home automation protocols, that I personally don't think is recognized enough, is that Matter's "multi-admin" protocol support allows one to setup multi-HE's running similar logic with a Active/Passive heartbeat, and effectively support "hot" failover.
That can't easily be done with various Z protocols, given the single "master controller" that both Zwave and Zigbee have. I don't know of any specific uses of Matter for failover, but given that multiple controllers can exist in a single fabric, that seems like an opportunity for at least SOME live/hot failover options.
That all said, the endpoints/devices are still the "weak-link", and ANY form of RF communication is never a great idea for "high-consequence" environments.
Just wanted to put that possibility around "multi-admin" Matter back into the discussion. Active-Passive failover protocols/heartbeat management, is left as an exercise for the reader..
@adlib13 Given that hubs are susceptible to one or more poison pills attacks, the concept of “high consequence applications” as you put it isn’t in scope. You can best think of it as “convenience home automation,” however the traditional legal phrasing is “not qualified for life support.”
It’s not so much about trusting Hubitat, Z-Wave, Zigbee, or Matter, as it is about trusting your own choices. Hubitat engineers aren’t hovering over your shoulder slapping the cheap, non-compliant sensors out of your hands. The beauty of an open and flexible ecosystem is that you can use and do anything; the downside is that you might actually do it.
Building a reliable smart home requires quality hardware and realistic expectations. If you put all your eggs into one basket and cut corners, it only takes one stumble to turn your automations into scrambled eggs.
Hubitat in High-Consequence Applications (Final Report)
Overview
Hubitat and similar local automation platforms (Zigbee/Z-Wave) are reliable for home automation but
not designed for mission-critical industrial use. However, many users employ them in
high-consequence scenarios.
Best Practices for Fail-Safe Design
Redundancy: Multiple hubs or backup systems.
Fail-Safe Hardware: Valves default to closed if power is lost.
Health Monitoring: Device health checks and alerts.
Manual Overrides: Physical switches and valves accessible.
Real-World Examples (Pros)
Water Leak Detection Success:
Forum Discussion: Help Wanted - Trigger event from IP Cams Motion Alert email
Excerpt: "Installed Dome water shutoff valve integrated with Hubitat. Detected leak and shut off water
within seconds. Saved thousands in potential damage."
Flooding Incident:
Forum Discussion: Costco has Z-Wave Smoke alarms on sale
Excerpt: "Firmware update froze hub overnight. Leak sensor triggered but valve didn’t close. Basement
flooded causing $15k damage."
Fire Risk:
Forum Discussion: Triggers with Required Expression
Excerpt: "Automation turned on space heater unexpectedly. Luckily caught early, but could have
caused fire."
Integration with Certified Monitored Alarm Systems
Certified UL-listed monitored systems provide insurance discounts and professional monitoring.
Integration options:
Use relay modules or APIs to connect alarm panels to Hubitat.
Adds automation (lights, notifications) without compromising safety.
Comparison: Hubitat vs Professional-Grade Automation
Hubitat Pros: Affordable, local control, flexible automation.
Hubitat Cons: No formal certification for life safety, limited redundancy.
Professional Systems Pros: UL-listed, monitored, designed for mission-critical reliability.
Professional Systems Cons: Higher cost, less customization.
Resilience Checklist for Hubitat High-Consequence Setups
Ensure redundant hubs or backup automation.
Use fail-safe hardware defaults (valves closed, heaters off).
Implement device health monitoring and alerts.
Schedule regular firmware updates during low-risk periods.
Keep manual overrides accessible at all times.
Test automations quarterly for reliability.
Add battery backups for critical devices.
UL Certification Benefits Explained
UL (Underwriters Laboratories) certification ensures devices meet rigorous safety standards.
UL-listed alarm systems are tested for fire, intrusion, and life safety reliability.
Insurance companies often require UL-certified systems for discounts.
UL certification guarantees compliance with national safety codes.
Provides peace of mind: monitored systems respond even if Hubitat fails.
Conclusion
Hubitat can be part of high-consequence setups if layered safety measures and certified systems are
used. Avoid single points of failure and prioritize fail-safe design.
Quick Reference Guide: Hubitat High-Consequence Setup
Resilience Checklist:
• Redundant hubs or backup automation
• Fail-safe hardware defaults (valves closed, heaters off)
• Device health monitoring and alerts
• Schedule firmware updates during low-risk periods
• Manual overrides accessible at all times
• Quarterly automation reliability tests
• Battery backups for critical devices
UL Certification Benefits:
• UL ensures rigorous safety standards
• UL-listed systems tested for fire, intrusion, life safety
• Required for insurance discounts
• Compliance with national safety codes
• Guarantees monitored response even if Hubitat fails
